user1 | Hello. I'm trying to create a init script for wireguard, but it seems it's not behaving as expected. ie, when i boot with no network, wg-quick tries to establish a connexion, blocking the boot process for like 1min | 00:06 |
---|---|---|
user1 | that's the init script im using: http://sprunge.us/lnG5vZ | 00:06 |
user1 | if anyone could point me where the issue lies/how to fix it, i'd be grateful :> | 00:07 |
mason | user1: Do you just want it to start manually? | 00:35 |
mason | If so, maybe: update-rc.d wireguard disable | 00:35 |
mason | Then you can "service wireguard start" and "service wireguard stop" at your convenience. | 00:36 |
user1 | i was more expecting it not to start if requirements aren't met (like, no network) | 00:37 |
user1 | but im not sure this can be accomplished. your solution might be more adequate | 00:37 |
mason | user1: You can inject some scaffolding to check for a working network, and some logic to decide what to do if there's no network, etc. Manual control sounds more pleasant to me. | 00:39 |
user1 | mh indeed | 00:42 |
user1 | thanks for the hints | 00:42 |
openbsdtai123 | hello a little question, on devuan ascii live, what is the default desktop 1) XFCE4 or 2) LXDE ? | 00:47 |
openbsdtai123 | thank you | 00:47 |
fsmithred | openbsdtai123, xfce | 00:53 |
openbsdtai123 | is autofs or usbmount preinstalled on this live? it seems that the auto mount works fine... very fine. | 00:57 |
openbsdtai123 | thank you | 00:58 |
user1 | (is there any specific chan regarding the chimaera release ?) | 00:59 |
fsmithred | usb is not mounted automatically - you have to click on the icon that pops up | 01:00 |
fsmithred | autofs is not installed | 01:01 |
fsmithred | user1, no, and chimaera is not yet a release | 01:01 |
fsmithred | be ready for dealing with breakage if you use it | 01:02 |
user1 | fsmithred: yes indeed, i canoot upgrade my system because of broken deps | 01:07 |
user1 | i'd be glad to help tho. even if im not that tech savvy :x | 01:07 |
systemdlete | so... any thoughts on xfce+policykit? | 01:11 |
fsmithred | not any that are fit to print | 01:16 |
systemdlete | :) | 01:17 |
systemdlete | fsmithred: Is this still an issue on Beowulf? | 01:17 |
systemdlete | I'm trying to justify upgrading from ascii | 01:18 |
systemdlete | or, maybe I just need to move over to mate | 01:18 |
systemdlete | (I like mate anyway) | 01:18 |
fsmithred | if you want shutdown/reboot buttons to work, you need policykit-1-gnome and all it pulls in | 01:22 |
fsmithred | you can exclude Recommends if you want | 01:23 |
fsmithred | systemdlete, I've been using beowulf since the beginning of the year. xfce works fine | 01:23 |
systemdlete | what happens if you try to launch, say, gufw? | 01:24 |
fsmithred | no clue | 01:24 |
systemdlete | (or another pk program | 01:24 |
systemdlete | xfce works fine here also) | 01:24 |
systemdlete | It's the pk stuff that seems to be iffy, judging by user reports I've found by searching the web | 01:25 |
fsmithred | is gufw a graphical program that you need to run as root? | 01:25 |
systemdlete | graphical ufw (uncomplicated firewall) | 01:25 |
systemdlete | It works fine on my host mate desktop, but not in my xfce VM | 01:25 |
fsmithred | well, if the authors or maintainers made a good pkexec file, it should work fine | 01:26 |
fsmithred | why not in xfce? | 01:26 |
systemdlete | it tells me I can't run as root, and then it tells me it is reporting me to the principal | 01:26 |
fsmithred | is it beowulf? | 01:26 |
systemdlete | ascii, which was the basis of my question | 01:26 |
systemdlete | both are ascii --host and VM | 01:27 |
systemdlete | but the host runs mate, whereas the guest runs xfce | 01:27 |
fsmithred | ok, so the problem is not due to changes in su | 01:27 |
fsmithred | can you run other programs as root on the desktop? | 01:27 |
systemdlete | (sorry if I was not clear) | 01:27 |
systemdlete | example? | 01:27 |
fsmithred | like a graphical text editor or file manager? | 01:27 |
systemdlete | why would I run a text editor as root? | 01:28 |
fsmithred | ok, I'm lost | 01:28 |
systemdlete | or do you mean another pk program? | 01:28 |
fsmithred | what's the problem? | 01:28 |
systemdlete | gufw requires a policykit login | 01:28 |
fsmithred | you have to be root to run a firewall, right? | 01:28 |
systemdlete | to make changes to the firewall, yes | 01:28 |
fsmithred | is there a gufw-pkexec or something like that to start it? | 01:29 |
systemdlete | must be | 01:29 |
fsmithred | well, you're running ascii | 01:29 |
fsmithred | how does it work | 01:29 |
fsmithred | ? | 01:30 |
fsmithred | or maybe it doesn't work? | 01:30 |
systemdlete | yes, there is a gufw-pkexec on both the host and the guest. | 01:30 |
fsmithred | and you have all the right 'kit shit installed in the vm? | 01:31 |
systemdlete | it works fine when I use the mate desktop on the host. I can give the root login when it prompts me and I get the GUI. But not in the guest where I am running xfce | 01:31 |
systemdlete | what would "all" of that be? | 01:31 |
systemdlete | (I'm no expert on this stuff) | 01:31 |
fsmithred | any other pkexec stuff work? synaptic or gparted? | 01:32 |
systemdlete | ah, gparted. I could try that. Hold on. | 01:32 |
systemdlete | well, gparted tells me you have to be root to run it (like, duh). If I run it from the menu, nothing happens at all. It just fades away. From user command line | 01:35 |
systemdlete | it tells me that and just exits | 01:36 |
systemdlete | when I click the OK button | 01:36 |
systemdlete | (sorry, fingers a bit unsteady here) | 01:36 |
systemdlete | I'll try synaptic | 01:36 |
systemdlete | http://paste.debian.net/1160978/ (similar output for gufw, different output for gparted though) | 01:38 |
systemdlete | fsmithred: Have you tried synaptic on xfce on beowulf? | 01:40 |
fsmithred | yes | 01:40 |
fsmithred | it works | 01:40 |
systemdlete | you are able to enter the root password and it takes it without an error? | 01:40 |
fsmithred | am I here? | 01:51 |
systemdlete | yes | 01:51 |
systemdlete | am I? | 01:51 |
fsmithred | yeah | 01:51 |
fsmithred | thunderstorm here | 01:51 |
systemdlete | oh | 01:51 |
systemdlete | I thought maybe you didn't like my question. LOL | 01:51 |
fsmithred | policykit-1-gnome elogind libpam-elogind | 01:52 |
systemdlete | Good to hear you are getting some wet there. We got a little here (Sacramento) a few days ago, but not since. | 01:52 |
fsmithred | we just got dumped on. Sorry I can't send you some. | 01:52 |
systemdlete | where are you? | 01:52 |
fsmithred | massachusetts | 01:52 |
systemdlete | oh, is that one of the hurricanes/tropical storms? | 01:53 |
fsmithred | not sure if they've arrived yet | 01:53 |
systemdlete | so those are the 3 packages I need then? | 01:53 |
fsmithred | and some libpolkits that come with them | 01:54 |
systemdlete | let me check... | 01:54 |
systemdlete | 234.4-2 for elogind and libpam-elogind | 01:55 |
systemdlete | do you have an ascii there with xfce? maybe you could try it and we could compare? | 01:56 |
systemdlete | if it fails for you also, then I have my answer. | 01:56 |
fsmithred | not one I can get to easily | 01:56 |
systemdlete | ok. | 01:56 |
fsmithred | but I know it works. I've spent time on this. Look on the forum. | 01:57 |
systemdlete | well, maybe someone else here has one. | 01:57 |
fsmithred | pkit-1-gnome | 01:57 |
systemdlete | ok | 01:57 |
fsmithred | systemdlete, how did you install the ascii? | 02:00 |
systemdlete | hmmm. | 02:00 |
systemdlete | thinking | 02:00 |
fsmithred | look in /var/log | 02:00 |
systemdlete | dpkg.log? | 02:01 |
fsmithred | refractainstaller.log? | 02:01 |
systemdlete | nope | 02:01 |
fsmithred | or maybe just installer if you used d-i | 02:01 |
systemdlete | I think it was a netinstall, though | 02:01 |
systemdlete | I see I have 2 devuan ISOs where I normally keep them | 02:02 |
systemdlete | both are netinstall, one for ascii and one for beowulf | 02:02 |
systemdlete | I found the installer subdir | 02:03 |
fsmithred | yup | 02:03 |
fsmithred | is the VM used for any real work? | 02:04 |
systemdlete | oh yeah. It's my "main" VM, where I do my personal accounting, job searches, church stuff, etc | 02:04 |
fsmithred | oh | 02:04 |
systemdlete | it's not a test VM, but I have tons of those | 02:05 |
fsmithred | make another one, about 12G and install beowulf to look at it | 02:05 |
fsmithred | ok | 02:05 |
fsmithred | try it, you'll like it | 02:05 |
systemdlete | I have a beowulf vm already! | 02:05 |
systemdlete | I use it for testing, and I have poked around in it a bit | 02:05 |
systemdlete | If/when I run into issues with beowulf, will the solution be to upgrade to Chimaera? :D | 02:07 |
fsmithred | no | 02:07 |
fsmithred | not for another year, at least | 02:07 |
systemdlete | (was j/k there, fsmithred) | 02:08 |
systemdlete | I searched the forum for policykit-1-gnome ascii xfce and got 4 hits | 02:08 |
systemdlete | none look related to my issue | 02:08 |
fsmithred | did gparted work or not? | 02:09 |
systemdlete | no | 02:09 |
fsmithred | did you try starting it from a root terminal? | 02:09 |
systemdlete | it gives me a popup telling me that only root can do this, and then I click a control and it goes away, never to be seen again | 02:10 |
systemdlete | (until I try it again, that is) | 02:10 |
systemdlete | I can make you a movie. | 02:10 |
fsmithred | what command are you using? | 02:11 |
systemdlete | I tried it from the menu; nothing happens, as I said before. | 02:11 |
systemdlete | I tried it from user command line -- | 02:11 |
systemdlete | that's where it gives the popup telling me mere plebes like me are not permitted to run gparted | 02:11 |
fsmithred | gparted-pkexec as user should give you a text dialog that asks for password | 02:11 |
fsmithred | which command? | 02:12 |
systemdlete | (I'd like to remind linux that it is MY computer...) | 02:12 |
fsmithred | su | 02:12 |
systemdlete | I ran gparted | 02:12 |
fsmithred | give root password | 02:12 |
fsmithred | gparted | 02:12 |
fsmithred | it should start | 02:12 |
systemdlete | I can run any of them as root, setting the DISPLAY to :0. yeah, then it works. | 02:12 |
systemdlete | but that's a rather finicky workaround | 02:12 |
fsmithred | in ascii, that should just work | 02:13 |
fsmithred | no display messing | 02:13 |
fsmithred | in beowulf, su is different | 02:13 |
systemdlete | I didn't try it without. I just assumed that root would not have access to the DISPLAY | 02:13 |
systemdlete | oh | 02:13 |
systemdlete | wait, you mean just plain "su" -- I always use "su -" to get a root env | 02:14 |
fsmithred | yes, just 'su' | 02:14 |
fsmithred | and you can use user's X | 02:14 |
systemdlete | sure, if I ran just plain su I wouldn't need the DISPLAY variable set | 02:14 |
systemdlete | but that's minor. | 02:14 |
systemdlete | the point is, I have to su to run a GUI program. | 02:14 |
systemdlete | Sounds like I need to plan a couple of upgrades here... | 02:14 |
fsmithred | either that or make a working pkexec file | 02:14 |
systemdlete | where are those kept? | 02:15 |
fsmithred | in /usr/share/polkit-1/actions | 02:17 |
fsmithred | good luck | 02:17 |
systemdlete | http://paste.debian.net/1160981/; it's the stock file that came with ascii | 02:18 |
fsmithred | that looks vaguely familiar | 02:19 |
systemdlete | http://paste.debian.net/1160983/ is an excerpt from the auth.log | 02:23 |
systemdlete | I think that's several tries. | 02:24 |
fsmithred | what display manager? | 02:24 |
systemdlete | lightdm | 02:26 |
systemdlete | is that it? | 02:26 |
systemdlete | istr there was a mismatch between dm's and window managers | 02:26 |
fsmithred | I really don't know | 02:27 |
fsmithred | you can try the fix for lightdm that's in the beowulf release notes | 02:27 |
fsmithred | that's specifically for the shutdown buttons on the login screen, but it might help with the session overall | 02:28 |
systemdlete | hmmm. ok | 02:28 |
fsmithred | this is a guess. I don't understand policykit at all. | 02:28 |
fsmithred | I understand that it's a major pain in the ass for me and a lot of other people. | 02:28 |
systemdlete | Along with SELinux, ACL's, and a zillion other bits of security tossed into the mix | 02:29 |
systemdlete | SELinux doesn't bite as much these days. They seem to have gotten the karma for various files just right. | 02:30 |
systemdlete | But every once in a while, I find that SELinux has to be addressed. | 02:30 |
systemdlete | I'll try the Beowulf release notes | 02:31 |
systemdlete | thanks again for your patience. | 02:31 |
fsmithred | good luck | 02:32 |
systemdlete | console-kit and elogind are both running. If they are mutually exclusive, why doesn't apt handle that? | 02:41 |
systemdlete | the release notes say to run "xprop --root --remove AT_SPI_BUS" but that throws an error. | 02:42 |
systemdlete | of course, I am on ascii, not beowulf. | 02:42 |
systemdlete | I did make the suggested change for the greeter though. | 02:42 |
fsmithred | not that | 02:47 |
fsmithred | in /etc/pam.d/lightdm | 02:47 |
systemdlete | well, the problem is gone, thanks! | 02:47 |
pablocastellanos | Tried for hours without success. It is possible to add nfs shares in /etc/fstab? With debian 6, debian 7, it worked. Tried in beowulf, and strange things happens at boot time. | 02:47 |
systemdlete | I completely obliterated console kit from my system, rebooted, and voila! | 02:47 |
systemdlete | I can successfully invoke gparted, etc. | 02:48 |
fsmithred | oh | 02:48 |
systemdlete | yeah, oh. | 02:48 |
systemdlete | LOL | 02:48 |
fsmithred | yeah, elogind or consolekit, not both | 02:48 |
systemdlete | apt really should enforce that | 02:48 |
systemdlete | I know I am pretty effusive with thanks, but thanks again. These little things pile up and turn into major annoyances. It is good to know I can come here to get some answers. | 02:50 |
systemdlete | I'll still plan on a beowulf upgrade soon. | 02:50 |
pablocastellanos | Looking at /etc/rcS.d/ supposedly mountall must not try to mount network filesystem, only local ones. then only after networking and nfs-common, mountnfs must mount nfs shares, but that's not happening. mountall tries to mount nfs shares, but rpcbind is not running, and errors appears at boot, and nfs shares are never mounted. | 02:51 |
fsmithred | do read the upgrade guide | 02:51 |
pablocastellanos | This is a brand new installation of beowulf, It is possible a bug on initscripts package? | 02:53 |
systemdlete | I have an ascii install where I have an nfs mount in the /etc/fstab. It works. | 02:55 |
pablocastellanos | systemdlete: I'll try to configure my nfs mounts in ascii then, upgrade to beowulf to see what happens. | 02:57 |
systemdlete | so you have an nfs entry in /etc/fstab and you have problems with this? | 02:58 |
fsmithred | I use autofs with my nfs | 02:58 |
systemdlete | on beowulf, I mean. | 02:58 |
pablocastellanos | systemdlete: Yes, it never mounts. | 02:58 |
pablocastellanos | systemdlete: A race condition occurs | 02:59 |
systemdlete | I'm not recommending you install ascii then upgrade to solve this. I doubt that would solve the problem. | 02:59 |
fsmithred | pablocastellanos, did you see if there's a bug report either in devuan or debian? | 02:59 |
systemdlete | which init are you using on beowulf? | 02:59 |
pablocastellanos | fsmithred: sysvinit | 03:00 |
pablocastellanos | fsmithred: Tried to install openrc, it happens too | 03:01 |
pablocastellanos | fsmithred: I'll check if there is any bug report | 03:01 |
rrq | pablocastellanos: perhaps adding a "_netdev" option to the nfs entries in fstab make some difference? | 03:01 |
fsmithred | are server and client both beowulf? | 03:02 |
rrq | mountall.sh mounting avoids those | 03:02 |
rrq | (and i'm not sure about the underscore) | 03:03 |
pablocastellanos | rrq: OK, I'll try | 03:03 |
pablocastellanos | fsmithred: Nope, server is debian jessie, client beowulf, but a manual mount -t nfs -o vers=3 jessie_ip:/path beowulf-path works. It's a problem with the mountall script of initscripts | 03:05 |
fsmithred | ok, you solved the problem I was expecting | 03:05 |
pablocastellanos | fsmithred: Probably fstab syntax changed as rrq suggests | 03:06 |
fsmithred | _netdev is correct | 03:06 |
fsmithred | I'm looking at man mount | 03:06 |
adhoc | morning folks | 03:08 |
adhoc | can someone point me in the right direction? | 03:08 |
fsmithred | --> | 03:08 |
adhoc | apt-get -yu dist-upgrade is throwing the following; | 03:08 |
adhoc | E: Repository 'http://deb.devuan.org/merged beowulf InRelease' changed its 'Suite' value from 'testing' to 'stable' | 03:08 |
adhoc | N: This must be accepted explicitly before updates for this repository can be applied. See apt-secure(8) manpage for details. | 03:09 |
adhoc | thanks fsmithred | 03:09 |
fsmithred | apt update | 03:09 |
fsmithred | then say 'yes' when it asks | 03:09 |
adhoc | ok, that is new to me | 03:09 |
adhoc | =) | 03:09 |
fsmithred | because it is new | 03:10 |
adhoc | feature || bug | 03:10 |
fsmithred | https://devuan.org/os/install | 03:10 |
pablocastellanos | adhoc: If you prefer apt-get, then apt-get --allow-releaseinfo-change update | 03:10 |
fsmithred | check the upgrade guide for the tricky parts | 03:10 |
adhoc | thanks fsmithred | 03:11 |
adhoc | pablocastellanos: that will save me a whole lot of hassle across my VM build farm ;) | 03:11 |
pablocastellanos | adhoc: :D | 03:12 |
adhoc | had not thought that installing beowulf, and it flipping from testing to stable would be an issue. | 03:14 |
fsmithred | the isos got made right before the repo was switched, so yeah, it's a bug | 03:17 |
adhoc | such is life =) | 03:17 |
adhoc | glad there is a work around | 03:17 |
adhoc | and you folks are helpful | 03:17 |
fsmithred | apt needing you to allow it is the new feature | 03:17 |
fsmithred | we didn't do that part | 03:18 |
adhoc | did that come from upstream? | 03:18 |
fsmithred | yes | 03:18 |
fsmithred | should only run into it if you were running beowulf before it went stable | 03:18 |
adhoc | I had to upgrade to get access to various language libraries | 03:19 |
golinux | Doesn't anyone read documentation to solve upgrade/ install issues anymore? | 05:09 |
furrywolf | I think most people are now used to software distributions where, rather than errata, a new build would be released with a workaround already in it. | 05:33 |
markizano | <golinux> Doesn't anyone read documentation to solve upgrade/ install issues anymore? <- LoL - no... :'( | 06:02 |
golinux | Well . . . newsflash . . . it's annoying to have to answer the same question dozens of times when the answer is available on all devuan channels. | 06:04 |
adhoc | golinux: If you are refering to my question above, I did search for about half an hour last week when I hit this problem last week | 06:17 |
adhoc | I didn't find anything. | 06:17 |
adhoc | I had another search this morning to see if htere wasn an update. | 06:18 |
adhoc | So I asked here, as I assumed I was doing something wrong. | 06:18 |
adhoc | "--allow-releaseinfo-change" is not in the man page either | 06:21 |
adhoc | From the error message, apt-secure talks about insecure repositories but not the mechanism required to move from testing to stable. | 06:24 |
adhoc | golinux: perhaps it is on the channels because it is not sufficiently well described on the web yet. | 06:25 |
adhoc | Hopefully in time this documentation will be created and you will see a drop in folks confused about what happened ... | 06:25 |
golinux | adhoc: Not referring to anyone in particular. I don't even remember your question. I am old an grumpy . . . | 06:28 |
adhoc | golinux: so what is IRC for it not asking questions? | 06:33 |
golinux | That is true. But questions that are well documented and discussed else are . . . well. | 06:43 |
golinux | Not saying that was your case btw. | 06:44 |
systemdlete | whew! That was nasty. Entire ascii host (and VMs) froze up. http://paste.debian.net/1160999/ | 09:14 |
systemdlete | Note that before this part of the log, there were many dozens of those USB errors ("xhci_hcd 0000:02:00.0: WARN Event TRB for slot 2 ep 4 with no TDs queued?") | 09:15 |
systemdlete | I switched my two USB3 devices to USB2 ports to see if this alleviates that problem. They are only warnings apparently, but I'm wondering if they might reflect some kind of instability. | 09:16 |
systemdlete | Several of the VMs do use USB devices, including those 2 USB3 devices. | 09:16 |
kreyren | when we will get emacs-27 in devuan! | 09:43 |
kreyren | https://pkginfo.devuan.org/stage/ascii/ascii/emacs_46.1.html what | 09:44 |
kreyren | not even ceres has it yet :c | 09:45 |
Junicchi | sup | 13:41 |
Junicchi | upgraded to devuan bullseye | 13:41 |
Junicchi | pulseaudio doesnt start automatically | 13:41 |
Junicchi | i have to start it manuall using pulseaudio --start | 13:41 |
Junicchi | i can use xinitrc to do it manually but | 13:42 |
Junicchi | anyone has an idea why it broke? | 13:42 |
zatumil | autospawn disabled? https://www.devuan.org/os/documentation/install-guides/beowulf/Release_notes_beowulf_3.0.0.txt | 13:45 |
Junicchi | its enabled | 13:48 |
Junicchi | ok i did that manually, didn't see the one in client.conf.d | 14:00 |
Junicchi | now no prob | 14:00 |
ShorTie | never heard of 'devuan bullseye' .. :/~ | 14:05 |
fsmithred | chimaera=bullseye Junicchi be sure to use codenames in sources.list (chimaera) | 14:18 |
crashoverride | any standard place to put the env vars to be set at shell invokation time? | 14:29 |
crashoverride | there's /etc/environment, but that seems to be a PAM thing only. | 14:32 |
fsmithred | crashoverride, maybe /etc/environment.d/ | 14:32 |
fsmithred | I have two files ending in .conf | 14:32 |
crashoverride | fsmithred: sure, but I was hoping for a HOME dotfile. | 14:32 |
debdog | usually .bashrc | 14:33 |
crashoverride | terrible idea. | 14:33 |
crashoverride | I was only looking at /etc/environment to see if there was a per-user definition. | 14:33 |
fsmithred | yeah, .bashrc should work | 14:33 |
crashoverride | bashrc would never work. | 14:33 |
debdog | or .bash_profile, depending on the env var | 14:33 |
crashoverride | 1. I do not, ever, use bash | 14:33 |
crashoverride | 2. bash isn't even installed. | 14:33 |
debdog | there must be something similar for your shell, too? | 14:34 |
crashoverride | yeah, but I am trying to clean my .zshrc | 14:34 |
crashoverride | and besides, I don't always use zsh, so I'd like something more portable. | 14:34 |
crashoverride | but no, not .profile | 14:34 |
debdog | you can source any file from withtin .zshrc | 14:35 |
crashoverride | I was searching for a place to specifically, preferrably defined in a standard, declare env vars. | 14:35 |
crashoverride | I know I can source files from my shell RC. | 14:35 |
crashoverride | I'm already doing this for aliases and functions. | 14:35 |
crashoverride | ah well | 14:36 |
* debdog scratches head "not entirey certain I do understand you" | 14:36 | |
debdog | *erly | 14:36 |
crashoverride | okay, you see how .profile is defined as a standard file to execute when the shell is starting? | 14:37 |
crashoverride | well, I was hoping there were other places for different things | 14:37 |
crashoverride | but apparently, just profile. | 14:37 |
debdog | AFAIK the root is /etc/profile. everything is getting sourced from there | 14:39 |
debdog | AKA, anything there is 'standard' | 14:39 |
crashoverride | Well, no. | 14:39 |
crashoverride | POSIX clearly defines the existence of a .profile file. | 14:40 |
crashoverride | at least it mentions it, I'm currently searching for the exact specification; but I'm using a terrible search engine so... it will take some time. | 14:40 |
debdog | hehe. if you find something, I'll appreciate a link | 14:40 |
debdog | says: | 14:41 |
debdog | # /etc/profile: system-wide .profile file for the Bourne shell (sh(1)) | 14:41 |
debdog | # and Bourne compatible shells (bash(1), ksh(1), ash(1), ...). | 14:41 |
debdog | I have no clue how zsh handles things. | 14:42 |
crashoverride | Well, for now, I have https://pubs.opengroup.org/onlinepubs/9699919799/xrat/V4_xcu_chap02.html in which is said: ENV: [...] The file referred to by ENV differs from $HOME/.profile in that .profile is typically executed at session start-up, whereas the ENV file is executed at the beginning of each shell invocation. | 14:42 |
crashoverride | zsh isn't very different and it also tries to be POSIX | 14:43 |
crashoverride | but again, I want my setup to work everywhere, not just in zsh | 14:43 |
fsmithred | The /etc/security/pam_env.conf file specifies the environment variables to be set, unset or modified by pam_env(8). When | 14:43 |
fsmithred | someone logs in, this file is read and the environment variables are set according. | 14:43 |
crashoverride | I'll try setting ENV to ~/.envvars | 14:43 |
crashoverride | fsmithred: indeed, but that requires PAM. | 14:44 |
crashoverride | which I try very hard not to use; so end up usually not using. | 14:44 |
debdog | I think I'll try that tonight. waiting for a chance to have a closer look at zsh for some time now. | 14:44 |
crashoverride | zsh is great. | 14:44 |
crashoverride | so many great productivity tweaks | 14:45 |
fsmithred | then maybe use /etc/login.defs | 14:45 |
crashoverride | fsmithred: I'll look at it, thanks. | 14:45 |
debdog | I know. but there is another factor: /me is lazy | 14:45 |
crashoverride | debdog: honestly | 14:45 |
crashoverride | debdog: install zsh, clone oh-my-zsh and install it | 14:45 |
crashoverride | debdog: and use the theme "agnoster" | 14:45 |
crashoverride | start from there. | 14:45 |
crashoverride | you'll see, instantly better. | 14:45 |
* debdog reads up on that on https://ohmyz.sh/ | 14:46 | |
ShorTie | 'instantly' ?? | 14:47 |
ShorTie | sounds like alot of work to me | 14:47 |
fsmithred | I use zsh to start a build with live-sdk and I haven't noticed any benefits. | 14:52 |
fsmithred | but I only use a few commands. I know there are some nice features for scripting. | 14:52 |
crashoverride | ShorTie: it's literally "apt get install zsh && git clone https://github.com/ohmyzsh/ohmyzsh ~/.oh-my-zsh && sed 's,export ZSH=.*$,export ZSH='"$HOME"'/.oh-my-zsh,' .oh-my-zsh/templates/zshrc.zsh-template >> .zshrc | 14:58 |
crashoverride | " | 14:59 |
crashoverride | meh, I should have prefixed the two last files with ~/ to make it foolproof | 14:59 |
crashoverride | fsmithred: not for scripting no. | 14:59 |
crashoverride | fsmithred: for scripting I use ash, dash, ksh and such | 14:59 |
crashoverride | fsmithred: the WHOLE benefit of zsh is interactive use. | 15:00 |
crashoverride | executing folders to cd | 15:00 |
crashoverride | advanced globbing | 15:00 |
crashoverride | interactive completion and selection | 15:00 |
crashoverride | filename expansion | 15:00 |
crashoverride | fuzzy matching | 15:00 |
crashoverride | etc | 15:00 |
suavedandy | Hello, guys. | 16:38 |
crashoverride | hi | 16:39 |
suavedandy | Does lid switch suspend the system on SysVinit or I have nothing to worry about? | 16:40 |
suavedandy | Coming from Debian. | 16:40 |
fsmithred | suavedandy, it's not working by default in ascii or beowulf, but it does work in chimaera (current testing suite) | 16:50 |
suavedandy | And that's good for me. | 16:51 |
fsmithred | well, you can set it to suspend or not in the power manager | 16:51 |
fsmithred | but that setting works only in chimaera | 16:51 |
fsmithred | and I'm talking about xfce - I don't know about the other desktops | 16:51 |
suavedandy | I don't have a GUI anyway. | 16:52 |
fsmithred | lol | 16:52 |
crashoverride | POSIX is hella confusing IMO | 16:52 |
fsmithred | I think you're safe | 16:52 |
suavedandy | Does Devuan have a documentation or something? | 16:53 |
fsmithred | yeah some. What do you want to know about? | 16:53 |
crashoverride | how to refine kerosen. | 16:54 |
crashoverride | using only household equipment and items. | 16:54 |
fsmithred | https://files.devuan.org/devuan_beowulf/Release_notes.txt | 16:54 |
suavedandy | Ah, nothing. I forgot that the repos are listed on the site. | 16:54 |
suavedandy | Wanted to find the package repos. | 16:54 |
fsmithred | example lines for sources.list are at devuan.org, if you want to poke around the actual files, they're at pkgmaster.devuan.org (or any package mirror) | 16:55 |
suavedandy | Eh, I know how to edit my sources file. I have Nano ready. | 16:57 |
suavedandy | Thanks. | 16:57 |
mason | suavedandy: You can have the lid switch working with a modicum of configuration. | 16:59 |
mason | suavedandy: I don't use it myself, as I like to control sleep state manually, but half a sec and I'll give you the files you want to edit. | 16:59 |
suavedandy | The problem is that the system might go to sleep when a script is rolling. | 17:00 |
suavedandy | You can imagine what a disaster it would be to stop a script halfway through. | 17:01 |
mason | suavedandy: Oh, do you NOT want it to sleep? | 17:01 |
suavedandy | Of course. | 17:01 |
suavedandy | No way in my life. I have Windows PTSD from that. | 17:02 |
* ShorTie Thinkz, He wants to give it a No-Doze before shutting the lid | 17:02 | |
mason | suavedandy: Generally sleep will not cause problems for scripts, but you can make sure it doesn't happen. The common way to support sleep is the acpi-support package. If you have that installed, look at /etc/default/acpi-support. If you need to edit it, do so and restart acpid. | 17:02 |
suavedandy | Also, I have a habbit of closing the lid to have a coffee break or something while APT's rolling. | 17:04 |
mason | Yeah, you don't want to sleep then. I set up the same way, where lid close is just that, and I have to tell the system to pm-suspend manually. | 17:04 |
mason | suavedandy: If you observe sleep and it's not acpi-support doing it, grep -r acpi will probably turn up what's doing it. | 17:05 |
mason | fsmithred: ^ in case the discussion is useful. Just re-verified that it works in Beowulf. | 17:06 |
mason | Now I'm turning it off again. | 17:06 |
fsmithred | mason, thanks. I'll re-test. Are you in xfce? | 17:06 |
mason | fsmithred: No, xdm/openbox. | 17:06 |
mason | fsmithred: acpi-support is DE-agnostic | 17:06 |
fsmithred | oh, ok | 17:07 |
fsmithred | I think that works in ascii, too | 17:07 |
mason | Ah, were you talking explicitly about XFCE not doing its thing out of the box? Sorry for the confusion if so. | 17:07 |
fsmithred | yeah, just a gui setting | 17:07 |
mason | Ah, hrm. | 17:07 |
mason | Well. Hm. One of my kids has Beowulf and XFCE, but it's not a laptop. I can maybe test a bit. | 17:08 |
suavedandy | If sleep doesn't cause issues then I guess it would be nifty in wireless mode. | 17:08 |
fsmithred | I only have ascii and chimaera on the laptop | 17:08 |
fsmithred | but I can boot a usb and try beowulf | 17:08 |
mason | suavedandy: It doesn't cause issues in that scripts tend to pick up right where they left off. | 17:08 |
suavedandy | Also, hibernation option would help a ton with power saving. | 17:08 |
suavedandy | The one that uses the hard drive. | 17:09 |
suavedandy | That option I find quite useful. | 17:10 |
debdog | crashoverride: in case you have not found anything yet: ftp://ftp.fu-berlin.de/pub/unix/shells/zsh/FAQ section "3.2: In which startup file do I put...?" | 17:41 |
crashoverride | debdog: thanks! | 17:41 |
crashoverride | interesting read. | 17:41 |
debdog | it's an extensive FAQ, have just had a quick glance but some things make me think I am probably not a zsh-guy :/ | 17:42 |
crashoverride | why? | 17:44 |
suavedandy | I like how the repo for security updates is beowulf-security instead of beowulf/updates. | 17:45 |
debdog | I am not a developer, no clue about C and stuff | 17:46 |
suavedandy | Tho I wonder why the URL addresses before the repos end with "merged." | 17:46 |
fsmithred | because they are merged | 17:46 |
fsmithred | we merge the /devuan packages with those from debian | 17:47 |
* suavedandy confoozed | 17:47 | |
suavedandy | Ah. | 17:47 |
fsmithred | we don't keep a full repo | 17:47 |
crashoverride | fool repo* | 17:47 |
crashoverride | as in "confoozed" | 17:47 |
fsmithred | amprolla pulls the un-forked packages from debian, filters out the banned packages and merges those with the forked (devuan) packages | 17:50 |
fsmithred | I didn't say that exactly right, but you probably understand what I mean | 17:50 |
suavedandy | Banned packages? | 18:29 |
suavedandy | Amprolla? | 18:30 |
golinux | https://pkgmaster.devuan.org/bannedpackages.txt | 18:35 |
golinux | https://git.devuan.org/devuan/amprolla3 | 18:36 |
golinux | suavedandy: ^^^ | 18:36 |
nemo | wait. tomcat9 is a banned package?? | 18:40 |
nemo | wth | 18:40 |
nemo | it's just a java based application server! | 18:40 |
nemo | did someone break its init packaging? | 18:40 |
nemo | that's problematic for future of devuan on about 10 machines at work | 18:41 |
golinux | So fork and fix it | 18:41 |
nemo | *sigh* | 18:41 |
golinux | It's been on the list for quite a while | 18:41 |
nemo | that's kinda linux response to everything, and fair, but there's only so many things I can submit code to | 18:42 |
nemo | guess I should look into what the problem is and how hard it is to fix | 18:42 |
golinux | Set your priorities | 18:42 |
nemo | golinux: does devuan have a bug server for this? | 18:42 |
golinux | bugs.devuan.org | 18:42 |
nemo | golinux: yeah. and I'm going to guess priority for this one is pretty low. easier to just give up on devuan. | 18:42 |
nemo | ok. lemme see if there's a discussion of the tomcat thing | 18:43 |
golinux | Everything i life is a choice | 18:43 |
nemo | hm... no tomcat bug | 18:43 |
nemo | hm. no discussion anywhere it seems | 18:44 |
golinux | Search the devuan channels | 18:44 |
golinux | I know it's been discussed. | 18:44 |
nemo | hmmm no systemd dependency in gentoo tomcat9. that's optimistic | 18:47 |
nemo | so it's probably just debian packaging | 18:47 |
fsmithred | what tomcat package is missing? I thought someone was working on tomcat, but I don't see any tomcat with +devuan in the version | 18:47 |
golinux | nemo: Here you go: https://lists.dyne.org/lurker/message/20200116.174138.fff7c445.en.html | 18:47 |
nemo | heh. just missing init script. lol | 18:47 |
fsmithred | I suspect there's no systemd dep in the tomcat9 in devuan, either | 18:48 |
nemo | it's even in the package source 😃 | 18:48 |
nemo | "However, in bullseye, the initscript has been | 18:48 |
nemo | removed from the package source already. " | 18:48 |
nemo | lame | 18:48 |
nemo | I thought debian voted to not do crap like that | 18:48 |
golinux | Bitch to debian | 18:48 |
fsmithred | removed by debian or removed upstream? | 18:48 |
nemo | oh. good point | 18:48 |
* nemo checks website | 18:49 | |
fsmithred | and the vote basically says that they can support sysvinit if they want to | 18:49 |
golinux | Anyway not our doing | 18:49 |
fsmithred | and should accept patches when offered | 18:49 |
nemo | fsmithred: hm | 18:49 |
fsmithred | yeah, it's not mandatory | 18:50 |
fsmithred | I think this was one of those times when it was better not to ask | 18:50 |
nemo | fsmithred: *should accept* - so if I passed this dude's build on to upstream debian, that would possibly cause it to be accepted? | 18:50 |
fsmithred | yeah, if you have a fix to make something work with sysvinit, you can submit it and it might get accepted | 18:51 |
nemo | ah. the upstream tomcat binary release for linux doesn't seem to include any init stuff at all. so this must be all debian's doing | 18:51 |
fsmithred | which is a better solution than devuan having to fork another package | 18:52 |
nemo | fsmithred: well, the message golinux linked to indicates someone made a fix already | 18:52 |
nemo | fsmithred: so I could literally just link to theirs (and maybe make a patch file) and ask upstream to accept it | 18:52 |
nemo | maybe just need to plug tab A into slot B | 18:52 |
nemo | bonus indicates people still care | 18:52 |
ShorTie | gonna be virually imposible to get them to include it | 18:54 |
ShorTie | how are they gonna test it ?? | 18:54 |
fsmithred | that thread also says that the devuanized version was moved to beowulf, but I don't see it | 18:56 |
fsmithred | so it must have been updated by debian and not re-devuanized | 18:56 |
nemo | fsmithred: huh... I thought you guys applied overlay patches to debian stuff. so those patches need bitrot fixes? | 18:57 |
daniel-molina | hello, is it possible to browse the packages with a web browser? | 18:57 |
nemo | fsmithred: you'd think adding init scripts would have low rot | 18:57 |
nemo | just due to no overlap w/ the systemd stuff | 18:57 |
fsmithred | daniel-molina, there's pkginfo.devuan.org to see what packages are available | 18:57 |
suavedandy | golinux: Lol, systemd blocked. (≧◡≦) | 18:58 |
fsmithred | and there's git.devuan.org to see the devuan packages we forked from debian | 18:58 |
daniel-molina | thanks fsmithred | 18:58 |
nemo | ShorTie: I would rather hope they wouldn't reject init script patches just due to inability to test in debian - that would make a mockery of that vote | 18:58 |
fsmithred | nemo, any time a new version goes into debian repo, we need to fix that package again | 18:59 |
nemo | fsmithred: that makes sense, I just thought maybe the process was automated - pull debian source package, attempt to apply .diff - if it succeeds with no errors, package and proceed to testing | 18:59 |
fsmithred | I think it was stated that it's up to the patch submitter to make sure it works | 18:59 |
nemo | but I don't know much about it obv | 18:59 |
nemo | fsmithred: ok. but each new upstream version requires a new patch resubmission with devuan? | 19:01 |
daniel-molina | do you know where I can find documentation about writing sysvinit init scripts? | 19:02 |
nemo | anyway. at least it looks like problems w/ tomcat9 re not insurmountable, and we are ok on 8 for short-term. will just tuck this away as a "filing bugs w/ devuan/debian on a quiet day" todo item | 19:02 |
nemo | daniel-molina: https://wiki.debian.org/LSBInitScripts/ | 19:03 |
daniel-molina | thanks, nemo | 19:06 |
nemo | np | 19:10 |
openbsdtai123 | do you know maybe if thre is an emulator for "Sega Model 2" arcade? | 19:35 |
linux_n | Devuan didn't detect my wifi. Is there something that needs to be done for it to detect wifi? | 20:36 |
suavedandy | BTW, guys. Where's all the audio. Isn't ALSA, like, built into the kernel? | 20:37 |
mason | linux_n: You might need firmware, depending on what wifi it is. | 20:38 |
mason | suavedandy: apt install alsa-utils and use alsamixer to make sure you're not muted | 20:38 |
mason | suavedandy: And yes, ALSA is there out of the box, and Firefox at least will use it. | 20:39 |
suavedandy | Also, where's ACPI config again? | 20:39 |
linux_n | ok thanks | 20:40 |
mason | suavedandy: acpi-support, and the file I mentioned was /etc/default/acpi-support | 20:41 |
suavedandy | Also, why's my encrypted root busy when SysVinit tries to unmount it on shutdown? | 20:41 |
mason | suavedandy: Ah, that's a bug we haven't fixed. Easily fixed. | 20:42 |
suavedandy | Oof. | 20:42 |
suavedandy | The amount of stuff you did for SysVinit support is staggering regardless. | 20:42 |
mason | suavedandy: https://bpa.st/RUEA | 20:42 |
fsmithred | add beowulf-proposed-updates and install cryptsetup-modified-functions | 20:42 |
suavedandy | That's a very good job. Am impressed. | 20:43 |
fsmithred | suavedandy, ^^^ | 20:43 |
mason | Oh, ignore what I said. fsmithred's answer will be easier. | 20:43 |
fsmithred | https://pkgmaster.devuan.org/devuan/pool/main/c/cryptsetup-modified-functions/ | 20:44 |
fsmithred | or download the .deb and install with dpkg | 20:44 |
fsmithred | it's for beowulf only | 20:44 |
mason | I've got the ASCII version of the patch if anyone needs it. | 20:45 |
fsmithred | oh, good | 20:45 |
mason | The one I posted was Beowulf, though. For completeness. | 20:45 |
suavedandy | And why when I choose "Console productivity" when installing Devuan the installer can't finish installing packages | 20:47 |
fsmithred | alt-f4 to read some messages | 20:47 |
suavedandy | Is it because console productivity tools are not included with the DVD ISO? | 20:48 |
fsmithred | if package installation fails, repeat that step | 20:48 |
fsmithred | oh | 20:48 |
fsmithred | possibly. You didn't select a mirror? | 20:48 |
suavedandy | Nope. | 20:48 |
fsmithred | un-check that one item and add what you can after the initial install | 20:49 |
suavedandy | Linux distros act funny with my Wi-Fi, soft-blocking it so I have to unblock it. That's why I install from the USB directly. | 20:49 |
fsmithred | apt show task-console-productivity (for a list of the deps) | 20:49 |
fsmithred | not deps - Recommends | 20:50 |
suavedandy | Guys. How do I change the hostname? | 22:32 |
user1 | sed -i 's/oldhostname/newhostname/g' /etc/host{s,name} | 22:33 |
user1 | well | 22:33 |
suavedandy | Guys. How do I change the hostname? | 22:35 |
user1 | user1 | sed -i 's/oldhostname/newhostname/g' /etc/host{s,name} | 22:36 |
suavedandy | host(s.name)? | 22:37 |
user1 | {} not () | 22:37 |
suavedandy | host[s,name]? | 22:38 |
suavedandy | {} | 22:38 |
suavedandy | host{s,name} | 22:38 |
suavedandy | There. | 22:38 |
user1 | yes | 22:38 |
suavedandy | I have localhost as my hostname. I did an offline installation so I wasn't prompted for a host name. | 22:39 |
user1 | localhost is a cool hostname | 22:39 |
user1 | you'll probably also need to restart the network | 22:41 |
user1 | or executing /etc/init.d/hostname.sh might be enough | 22:42 |
user1 | i guess that's it. the sed stuff, and hostname.sh | 22:42 |
suavedandy | user1: Does "user1 |" mean anything? | 22:59 |
gnarface | suavedandy: i'm guessing he meant to flag you but flagged himself instead. "user1 |" couldn't be part of the command unless the binary "user1" is on your system in your path somewhere, which it's not unless you put it there yourself | 23:01 |
user1 | no, i copy pasted what i said earlier | 23:01 |
user1 | indeed that's not part of the command | 23:01 |
gnarface | ah, i see. most clients i've seen copy& paste names like this: <user1> | 23:02 |
suavedandy | "ip6-newhostname" | 23:03 |
suavedandy | Are you sure it was a good idea to replace localhost with a new hostname, User1 | 23:04 |
user1 | sorry ? | 23:06 |
user1 | if it didn't work, you'll have to manually edit the /etc/hosts and /etc/hostname files to apply the changes | 23:07 |
suavedandy | Eh, it's okay. I just inversed the hostnames in your command and deleted brackets and ",name" | 23:08 |
suavedandy | That should work, I guess. https://wiki.archlinux.org/index.php/Network_configuration#Local_hostname_resolution | 23:10 |
suavedandy | No. | 23:14 |
suavedandy | It's actually this. https://linuxhandbook.com/debian-change-hostname/ | 23:14 |
suavedandy | There we go. | 23:14 |
gnarface | suavedandy: you should read the /etc/hosts man page, it's not hard to handle customization. in most cases you want to define a hostname *and* localhost (ideally not pointing to the same address) | 23:27 |
suavedandy | Should have just searched for it before | 23:30 |
gnarface | yes, that's a very effective approach | 23:30 |
gnarface | in most cases where debian has replaced something with systemd, the old wheezy documentation still applies to devuan | 23:31 |
gnarface | just fyi | 23:31 |
Wafficus | Hi there, how do I change my date timezone on Devuan? | 23:32 |
Wafficus | * from Central to EDT | 23:32 |
Wafficus | i've done this on systemd based distros before, but not on this to be honest | 23:32 |
gnarface | Wafficus: should be the same as debian still, but i guess i don't know for sure: apt-get install tzdata && dpkg-reconfigure tzdata | 23:33 |
Wafficus | ok thanks | 23:33 |
gnarface | if they really replaced that with some systemd-specific garbage, that would be infuriating | 23:34 |
Wafficus | sam@samdevuan ~ $ dpkg-reconfigure tzdata | 23:34 |
Wafficus | -bash: dpkg-reconfigure: command not found | 23:34 |
suavedandy | gnarface: Yeah, the manpage basically stated the same as that Linux Handbook article but also suggested changing one digit in the IP address just like ArchWiki. | 23:35 |
Wafficus | I did make sure I had "tzdata" installed" | 23:35 |
gnarface | Wafficus: WTF did you do!?!?! | 23:35 |
Hurgotron | Wafficus: be root for dpkg-reconfigure | 23:36 |
gnarface | Wafficus: just kidding. still though, that's alarming that you're missing that. could be a bad sign. it's in the debconf package though. try installing that | 23:36 |
gnarface | debconf: /usr/sbin/dpkg-reconfigure | 23:36 |
Hurgotron | gnarface: notice his prompt :) | 23:37 |
Wafficus | debconf is already installed | 23:37 |
Wafficus | yeah now with root I'm getting the ncurses menu for tzdata | 23:38 |
gnarface | oh, right | 23:38 |
Wafficus | will changei t from here | 23:38 |
Wafficus | *it | 23:38 |
Wafficus | thanks for the headsup | 23:38 |
Wafficus | thanks that worked | 23:38 |
Wafficus | thanks a ton gnarface Hurgotron | 23:38 |
gnarface | Wafficus: no problem. you probably should know you can dpkg-reconfigure a lot of packages very helpfully | 23:39 |
gnarface | Wafficus: (notably, if you didn't know about this, you probably should try it on the "exim4-config" and "locales" packges, too) | 23:39 |
suavedandy | How is the console productivity metapackage called again? | 23:40 |
gnarface | suavedandy: not sure but guessing "task-console-productivity" | 23:40 |
Jjp137 | that's correct | 23:40 |
suavedandy | Thanks. | 23:40 |
gnarface | there's a bunch of them called task-* | 23:41 |
gnarface | they will show up in the searches | 23:41 |
gnarface | useful stuff like presets for laptops and such | 23:41 |
suavedandy | fbi package… | 23:42 |
suavedandy | Wuh? | 23:42 |
gnarface | hehe | 23:42 |
gnarface | fbi - Linux frame buffer image viewer | 23:42 |
suavedandy | Oh. | 23:42 |
gnarface | Frame Buffer Image | 23:42 |
gnarface | someone thinks they're clever or just wasn't from the US | 23:42 |
gnarface | i don't know | 23:42 |
suavedandy | Man, there's so much bloat in this metapackage. | 23:43 |
gnarface | well, that's kinda what they're for | 23:44 |
user1 | there's the kgb package as well | 23:44 |
gnarface | you could try adding "--no-install-recommends" to the command-line options, might slim it down, might not | 23:44 |
suavedandy | Why use parted instead of… I dunno, fdisk? | 23:44 |
gnarface | fdisk has no GPT support | 23:45 |
suavedandy | user1: Heh. | 23:45 |
gnarface | actually i think fdisk only technically partially supports the MSDOS partition table format even, but well enough to make MS operating systems happy with it | 23:45 |
gnarface | there are other partitioning tools that also support GPT but they're even more arcane than parted | 23:46 |
suavedandy | If I'll install parted, can I delete fdisk or some packages and scripts depend on it? | 23:46 |
gnarface | not sure honestly. i would suspect that nothing depends on fdisk but if something does it will sure tell you | 23:47 |
gnarface | what i am sure of is that fdisk takes up almost no space so your time is better spent culling elsewhere (texlive packages installed still? you probably don't need those) | 23:47 |
suavedandy | There's also libfdiskl. | 23:48 |
gnarface | hmmm, a surprising amount of things might actually need it now that i look | 23:49 |
gnarface | try this: apt-cache rdepends fdisk | 23:49 |
\0xc0ffee | There is a build of fdisk that supports GPT, it is at least true on Fedora/CentOS systems, unsure about debian-based though (and that's a whole other point of frustration) | 23:49 |
gnarface | \0xc0ffee: to be honest my information on that could be up to 5 years old (last time i struggled with converting an old GPT drive that had some disturbingly resistant windows7 cruft left over in the boot sectors) | 23:50 |
\0xc0ffee | A small pity the authors of 'parted' never included a means to specify the end of a partition by specifying how large of a partition you want, rather than making you break out the calculator | 23:50 |
gnarface | yea but i don't think sfdisk or cfdisk are any more helpful in that regard, are they? | 23:51 |
\0xc0ffee | fdisk let you specify the end of a partition by specifing a size, e.g. 50GB | 23:52 |
gnarface | well, now it does. within my professional career it could still only do block counts | 23:52 |
gnarface | i still run across versions like that here and there | 23:53 |
\0xc0ffee | parted is one such tool, it's something that frustrates me somewhat, as we could have both | 23:53 |
gnarface | well, you know what they say... if you dislike it enough, then fork it. if it's not upsetting enough to fork then maybe it's not a big deal. | 23:54 |
\0xc0ffee | As for 'disturbingly resistant cruft' - RAID metadata left over on drives salvaved from an old LSI Logic array were a considerable pain for me | 23:55 |
\0xc0ffee | gnarface: Unfortunately, forking it to fix it requires knowledge of programming; to assume that Linux users are fluent in C is perhaps unfair | 23:56 |
gnarface | \0xc0ffee: well, it might be considered unfair but it might also be considered tough love | 23:56 |
suavedandy | Do I put kernel module blacklists to /etc/modprobe.d/ as usual? The Almighty Beep has returned. And my parents are, like, sleeping… Jesus. | 23:56 |
\0xc0ffee | gnarface: So the question then becomes: Do you want Linux to be adopted and used by ordinary (non programmer) users, or do you feel it should be the exclusive domain of developers? | 23:57 |
suavedandy | That damn beep. I'll get a heart attack from it faster then I become a senior. | 23:57 |
\0xc0ffee | The latter is the path adopted wholesale by OpenBSD which makes no bones about that point and explicitly states that fact | 23:57 |
gnarface | \0xc0ffee: (a means by which to encourage more people to familiarize themselves with C... mind you i'm not putting any effort into adding GPT support to fdisk either, so i'm just as guilty of apathy and laziness here) | 23:57 |
Generated by irclog2html.py 2.17.0 by Marius Gedminas - find it at https://mg.pov.lt/irclog2html/!